I'm Karthik Nadar, a passionate Cloud & DevOps Engineer and Full-Stack Developer. I am dedicated to crafting scalable infrastructures and pushing the boundaries of machine learning.

With a solid foundation in building resilient backend architectures using Express.js, FastAPI, and Python, I strive to create seamless user experiences through expertise in frontend technologies like React and Next.js.

My proficiency in cloud technologies includes:

I pride myself on designing efficient CI/CD pipelines and utilizing orchestration tools to ensure robust and scalable infrastructure.

In the realm of machine learning, I specialize in:

• Deep Learning: Building and fine-tuning CNN models
• Generative AI: Exploring GANs, VAEs, energy-based models, and diffusion models
• Text-to-Image Models: Fine-tuning diffusion models on custom datasets using innovative methods like LORA and Dreambooth

My focus on generative AI allows me to develop and deploy solutions that have a tangible impact on real-world applications.

I specialize in creating AI agents using the Vercel AI SDK and tools for task automation, leveraging generative AI to build both practical and innovative applications. By integrating cutting-edge AI APIs, I craft intelligent systems capable of:

• Automating routine and complex tasks
• Delivering enhanced user interactions
• Creating highly productive and useful applications for real-world problems

My expertise in these areas allows me to deliver robust solutions that merge automation with user-friendly design.

Visit Plum AI
Developed an AI-powered search engine similar to Perplexity, leveraging the Tavily LLM Search API and OpenAI API. Plum AI delivers more accurate and context-aware search results, enabling users to search efficiently and gain insights with better accuracy than traditional search engines.

Visit KitKat AI
An intelligent note-taking app that enables users to create digital sticky notes for quick idea capture and organization. Leveraging conversational AI, KitKat AI allows users to chat with their notes, effortlessly retrieving and summarizing past entries based on natural language queries.

Visit Avocado AI
Developed a text-to-video project that uses Firebase Storage to manage audio and image assets, Google Text-to-Speech (TTS) for voice generation, and Remotion.js to generate videos based on the content’s emotional tone. The project integrates Clerk for user authentication and uses Drizzle ORM with a PostgreSQL database on Neon. It provides a seamless experience in converting text input into engaging video outputs.

Discover my work on GitHub, including:

• Backend Architectures: Resilient solutions using Express.js, FastAPI, and Python.
• Frontend Development: Seamless user experiences with React and Next.js.
• Cloud Infrastructure: Scalable solutions with AWS, Docker, and Kubernetes.
• Machine Learning Models: Innovative text-to-image solutions with stable diffusion models.

Connect with me on LinkedIn or explore my repositories to learn more about my work and collaborations.

Thank you for visiting my GitHub. Let's build something amazing together!